156 THE UNIVERSITY SCIENCE BULLETIN. 



Numerous observers have remarked on the antigenic differences 

 in typhoid. Durham (8) observed such differences, but did not 

 attempt to group his strains. Weiss (1) and Hooker (9), however, 

 offered a tentative grouping on the basis of their agglutination and 

 absorption tests. 



The agglutination tests in this series were all done with suspen- 

 sions in sterile saline made from twenty-four-hour cultures. The 

 serum used came principally from rabbits immunized in this labo- 

 ratory. 



A high-titred bivalent horse serum from the New York city board 

 of health* prepared from the Mt. Sinai strain, and a freshly isolated 

 strain as well as a high-titred serum for which the Rawlings strain 

 had been used for immunization from the Lederle laboratories, were 

 also used. Table IV gives a^ summary of the results. In addition 

 to the results given here, eight other immune sera were used for 

 agglutination against all the organisms with similar results. 



The following technique was used for the absorption tests: The 

 serum to be tested was diluted to one-tenth of the titre. This di- 

 lution was then saturated with organisms, washed from a twenty- 

 four-hour agar slant to make a heavy emulsion. This was incu- 

 bated at 37° C. for four hours and for four days at ice-box tempera- 

 ture, more organisms being added as the supernatant fluid became 

 clear. The control of diluted serum in every case gave a good ag- 

 glutination in spite of the prolonged incubation. If the control gave 

 agglutination after absorption with the homologous organism the 

 test was repeated. 



Since considerable prominence has been given to the mirror re- 

 action in the recent literature, it might be well to establish some 

 standard method for absorption tests in order to get comparable 

 results. We found the following points must be carefully considered 

 in any test : 



1. Weight of suspension. 4. Repeated saturation. 



2. Dilution of serum. 5. Temperature. 



3. Time of absorption. 6. Controls. 



Krumwiede (4) recommends a proportion of 1-4 or 3, or at most 

 1-2 of packed cells to supernatant fluid. Our proportion after the 

 final centrifugation was about 1-3. It was found that a dilution 

 of one-tenth the titre of the serum was perfectly satisfactory. Al- 

 though higher dilutions could be used, a lower dilution did not give 

 complete absorption. Three or four hours was not long enough 



* I am indebted to the kindness of Dr. Charles Krumwiede for the use of this serum. 



